How variability shapes learning and generalization
Open accessThe DOI matches its Crossref record; 34 of the 34 references it cites were found in a registry or an index.
Abstract
Learning is using past experiences to inform new behaviors and actions. Because all experiences are unique, learning always requires some generalization. An effective way of improving generalization is to expose learners to more variable (and thus often more representative) input. More variability tends to make initial learning more challenging, but eventually leads to more general and robust performance. This core principle has been repeatedly rediscovered and renamed in different domains (e.g., contextual diversity, desirable difficulties, variability of practice). Reviewing this basic result as it has been formulated in different domains allows us to identify key patterns, distinguish between different kinds of variability, discuss the roles of varying task-relevant versus irrelevant dimensions, and examine the effects of introducing variability at different points in training.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
Multilayer feedforward networks are universal approximators
Hornik, K., Stinchcombe, M., & White, H. (1989) · Neural Networks · doi:10.1016/0893-6080(89)90020-8 · Cited by 21,728
A survey on Image Data Augmentation for Deep Learning
Shorten, C., & Khoshgoftaar, T. M. (2019) · Journal of Big Data · doi:10.1186/s40537-019-0197-0 · Cited by 13,203 · cited 1×
“Learning that is overly speci fi c to the training set yields progressively poorer performance on new items (so-called over fi tting, Figure IA), limiting the usefulness of the algorithm to correctly respond only to items that very closely resemble those that had been presented in training [140].”
A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play
Silver, D., Hubert, T., Schrittwieser, J., Antonoglou, I., Lai, M., Guez, A., Lanctot, M., Sifre, L., Kumaran, D., Graepel, T., et al. (2018) · Science · doi:10.1126/science.aar6404 · Cited by 3,562 · cited 1×
“In the past 10 years, the increase in processing power, optimization of learning algorithms, and, perhaps most importantly, increase in the size of training sets, has increasingly produced super-human performance on tasks ranging from object classification [169], to face recognition [170], to language processing [171]…”
BackgroundSpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
Park, D. S., Chan, W., Zhang, Y., Chiu, C.-C., Zoph, B., Cubuk, E. D., & Le, Q. V. (2019) · Interspeech 2019 · doi:10.21437/interspeech.2019-2680 · Cited by 3,558
Context theory of classification learning
Medin, D. L., & Schaffer, M. M. (1978) · Psychological Review · doi:10.1037/0033-295x.85.3.207 · Cited by 3,061
The Effectiveness of Data Augmentation in Image Classification using Deep Learning
Perez, L., & Wang, J. (2017) · arXiv · doi:10.48550/arxiv.1712.04621 · Cited by 2,499
The echo chamber effect on social media
Cinelli, M., De Francisci Morales, G., Galeazzi, A., Quattrociocchi, W., & Starnini, M. (2021) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2023301118 · Cited by 2,092 · cited 1×
“Recent studies focusing on social networks formed by social media algorithms, which tend to pair users with like-minded users, show that doing so leads to echo chambers [187] with the power to further polarize people’s opinions [188].”
BackgroundLearning and development in neural networks: the importance of starting small
Elman, J. L. (1993) · Cognition · doi:10.1016/0010-0277(93)90058-4 · Cited by 1,988
Distributed practice in verbal recall tasks: A review and quantitative synthesis.
Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006) · Psychological Bulletin · doi:10.1037/0033-2909.132.3.354 · Cited by 1,888 · cited 1×
“massed) learning of items consistently shows benefits’ [101] [101,131–135]”
BackgroundMaturational constraints on language learning
Newport, E. L. (1990) · Cognitive Science · doi:10.1207/s15516709cog1401_2 · Cited by 1,738 · cited 1×
“These fi ndings are in line with the idea that learning may bene fi t from ‘ starting small ’ (i.e., that having less data or less complex data early on provides a learning advantage, e.g., better retention) [82,83].”
Schmidt, R. A., & Bjork, R. A. (1992) · Psychological Science · doi:10.1111/j.1467-9280.1992.tb00029.x · Cited by 1,582 · cited 1×
“Instead, the exact same principle appears under different terminology in a host of fi elds and their insights have not been uni fi ed into a single theoretical framework (but see [25,26] for attempts to link the literature on motor and/or verbal learning to pedagogical applications).”
The law of group polarization.
Sunstein, C. R. (2002) · Journal of Political Philosophy · doi:10.1111/1467-9760.00148 · Cited by 1,371
Infant sensitivity to distributional information can affect phonetic discrimination
Maye, J., Werker, J. F., & Gerken, L. (2002) · Cognition · doi:10.1016/s0010-0277(01)00157-3 · Cited by 1,364
Contextual interference effects on the acquisition, retention, and transfer of a motor skill.
Shea, J. B., & Morgan, R. L. (1979) · Journal of Experimental Psychology: Human Learning and Memory · doi:10.1037/0278-7393.5.2.179 · Cited by 1,227
Paas, F. G. W. C., & Van Merriënboer, J. J. G. (1994) · Journal of Educational Psychology · doi:10.1037/0022-0663.86.1.122 · Cited by 1,198 · cited 2×
“increased variability may also be expected to hinder learning’ [62]”
BackgroundA Distributed Representation of Temporal Context
Howard, M. W., & Kahana, M. J. (2002) · Journal of Mathematical Psychology · doi:10.1006/jmps.2001.1388 · Cited by 1,174
Memory and Metamemory Considerations in the Training of Human Beings
Bjork, R. A. (1994) · Metacognition · doi:10.7551/mitpress/4561.003.0011 · Cited by 1,084
Lively, S. E., Logan, J. S., & Pisoni, D. B. (1993) · The Journal of the Acoustical Society of America · doi:10.1121/1.408177 · Cited by 788 · cited 1×
“In contrast, listeners trained with only a single talker showed little evidence of generalization to new tokens or new talkers’ [122] ‘When infants are exposed to variable exemplars of words, their learning is focused on the consistent pieces of information - in this case, phonological information.”
A review of the contextual interference effect in motor skill acquisition
Magill, R. A., & Hall, K. G. (1990) · Human Movement Science · doi:10.1016/0167-9457(90)90005-x · Cited by 701 · cited 4×
“High contextual interference, even though causing immediate limited performance, leads to superior performance on retention and transfer tests ’ [27] [27 – 31, 71, 110 – 114] Language Phonetic”
InfluentialA meta-analytic review of the distribution of practice effect: Now you see it, now you don’t
Donovan, J. J., & Radosevich, D. J. (1999) · Journal of Applied Psychology · doi:10.1037/0021-9010.84.5.795 · Cited by 696
On interpreting the effects of repetition: Solving a problem versus remembering a solution
Jacoby, L. L. (1978) · Journal of Verbal Learning and Verbal Behavior · doi:10.1016/s0022-5371(78)90393-6 · Cited by 667
Spacing Effects in Learning: A Temporal Ridgeline of Optimal Retention
Cepeda, N. J., Vul, E., Rohrer, D., Wixted, J. T., & Pashler, H. (2008) · Psychological Science · doi:10.1111/j.1467-9280.2008.02209.x · Cited by 631
Learning Concepts and Categories: Is Spacing the “Enemy of Induction”?
Kornell, N., & Bjork, R. A. (2008) · Psychological Science · doi:10.1111/j.1467-9280.2008.02127.x · Cited by 602
A symbolic-connectionist theory of relational inference and generalization.
Hummel, J. E., & Holyoak, K. J. (2003) · Psychological Review · doi:10.1037/0033-295x.110.2.220 · Cited by 567 · cited 1×
“See [151 – 167].”
Component-levels theory of the effects of spacing of repetitions on recall and recognition
Glenberg, A. M. (1979) · Memory & Cognition · doi:10.3758/bf03197590 · Cited by 444 · cited 2×
“…with the encoding variability theory, which suggests that more time between training blocks creates a greater opportunity for general, contextual, and descriptive cues to change, increasing the likelihood that an item at test would be similar to one of the items seen during training [67,68].”
Data Augmentation for Deep Neural Network Acoustic Modeling
Cui, X., Goel, V., & Kingsbury, B. (2015) · IEEE/ACM Transactions on Audio, Speech, and Language Processing · doi:10.1109/taslp.2015.2438544 · Cited by 376
Learning to Segment Speech Using Multiple Cues: A Connectionist Model
Christiansen, M. H., Allen, J., & Seidenberg, M. S. (1998) · Language and Cognitive Processes · doi:10.1080/016909698386528 · Cited by 356
Janiszewski, C., Noel, H., & Sawyer, A. G. (2003) · Journal of Consumer Research · doi:10.1086/374692 · Cited by 268
Statistical theory of distributional phenomena in learning
Estes, W. K. (1955) · Psychological Review · doi:10.1037/h0046888 · Cited by 246 · cited 2×
“categorization Distribution in time [66] X X X X X X X X Broader coverage”
The spacing effect in children’s memory and category induction
Vlach, H. A., Sandhofer, C. M., & Kornell, N. (2008) · Cognition · doi:10.1016/j.cognition.2008.07.013 · Cited by 165 · cited 3×
“…training (which are considered to be more variable) lead to better learning and broader transfer of motor skills (e.g., volleyball serves [27,28], badminton serves [29], golf strokes [30], pistol shooting [31]) and of novel categories (e.g., trivia facts [32], toys [33], scienti fi c concepts [34]).”
Spacing repetitions over 1 week
Glenberg, A. M., & Lehmann, T. S. (1980) · Memory & Cognition · doi:10.3758/bf03213772 · Cited by 151
Hebert, E. P., Landin, D., & Solmon, M. A. (1996) · Research Quarterly for Exercise and Sport · doi:10.1080/02701367.1996.10607925 · Cited by 133
Environmental drivers of systematicity and generalization in a situated agent
Hill, F., Lampinen, A., Schneider, R., Clark, S., Botvinick, M., McClelland, J. L., & Santoro, A. (2019) · arXiv · doi:10.48550/arxiv.1910.00571 · Cited by 81 · cited 2×
“…the general phenomenon in all of these examples is essentially the same: greater variability may initially hinder learning, but typically leads to an improved ability to generalize learning to new contexts by facilitating the formation of more abstract knowledge (e.g., [4,15 – 17]; Figure 1).”
Background
Cited by
Czyż, S. H., Wójcik, A. M., & Solarská, P. (2024) · Frontiers in Psychology · doi:10.3389/fpsyg.2024.1377122 · Cited by 10
“One could argue that the ultimate goal of practice is to transfer skills from one context to another, i.e., generalize practiced skill to a non-practice one (Raviv et al., 2022).”
BackgroundUnfamiliar face matching ability predicts the slope of face learning
Baker, K. A., & Mondloch, C. J. (2023) · Scientific Reports · doi:10.1038/s41598-023-32244-w · Cited by 6
Metadata
- Authors
- Raviv, L., Lupyan, G., & Green, S. C.
- Venue
- Trends in Cognitive Sciences
- Publisher
- Elsevier BV
- Published
- 2022-06
- Volume
- 26
- Issue
- 6
- Pages
- 462-483
- ISSN
- 1364-6613 (ISSN-L)
- OpenAlex
- W4280547462
- Cited by
- 206
- References
- 34
- FWCI
- 5.41
- Type
- journal article
- Language
- en
- Open access
- bronze
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.